Вы находитесь на странице: 1из 12

JO U R N A L O F

INTERNET LAW

VOLUME 12
NUMBER 6

DECEMBER 2008

EDITED BY

DLA

PIPER

Transcending Net Neutrality: Ten Steps


Toward an Open Internet
Sascha D. Meinrath and Victor
W. Pickard
he past few years have witnessed a once-obscure issue
known as net neutrality blow up into arguably
the most publicized policy debate in US telecommunications history. An untold story is how this
relatively technical debate spilled outside the rarefied
airs of Congressional Committees and the Federal
Communications Commissions (FCCs) eighth floor to
rage across the blogosphere, major newspapers, YouTube
clips, and episodes of The Daily Show to become, if not
a household phrase, a topic of popular debate involving
millions of Americans. One explanation is that, at its
root, the net neutrality debate is far more significant
than a squabble among technocrats. Rather, it is first
and foremost a normative debate, one that will determine the role of the Internet in a democratic society,
with profound implications for the daily welfare of millions of citizens who rely on the Internet as a critical
resource. Unfortunately, it is such normative concerns,
along with related political and historical contexts, that
have been least explored in much of the net neutrality
scholarship to date. This article aims to address these
gaps while expanding the parameters of the existing
debate.

Sascha D. Meinrath is the Research Director for the New America


Foundations Wireless Future Program. Additionally, he coordinates
the Open Source Wireless Coalition, a global partnership of open
source wireless integrators, researchers, implementers, and companies
dedicated to the development of open source, interoperable, low-cost
wireless technologies. He can be reached at sascha@saschameinrath.
com and (202) 986-2700 x226. Victor W. Pickard recently finished his
PhD at the Institute of Communications Research at the University of
Illinois, Urbana-Champaign. Currently, he works on telecommunications
policy as a Research Fellow for the New America Foundation.

Network neutrality, defined broadly, is nondiscriminatory interconnectedness among data communication networks that allows users to access the content
and to run the services, applications, and devices of their
choice. In essence, network neutrality forbids preferential treatment of specific content, services, applications,
and devices that can be integrated into the network
infrastructure. This principle has been the foundation
for rapid innovation and the Internets relative openness. As Congress debates whether network neutrality
protections should be written into current legislation,
the battle lines have been drawn between large telecommunications companies that own the pipes, on one
side, and Internet content companies and public interest
groups on the other. Although scholarship has begun to
catch up with the net neutrality debate, the majority
of this work has failed to connect this issue with larger

Continued on page 12

IN THIS ISSUE
TRANSCENDING NET NEUTRALITY: TEN STEPS
TOWARD AN OPEN INTERNET . . . . . . . . . . . . . . . . . . . . . 1
By Sascha D. Meinrath and Victor W. Pickard
ANTI-SOCIAL NETWORKING: LEARNING
THE ART OF MAKING ENEMIES IN WEB 2.0 . . . . . . . . . . 3
By Sajai Singh
FEDERAL LEGISLATIVE UPDATE . . . . . . . . . . . . . . . . . . . 22
By Bennet Kelley

J O U R N A L O F I N T E R N E T L AW

Transcending Net Neutrality


Continued from page 1
ramifications that arise while striving toward a democratic
Internet.
In this article, we attempt to broaden the net neutrality debate while briefly taking stock of recent scholarship.
In doing so, we critically evaluate the current network
neutrality debate and offer a set of technical and policy
guidelines for a new, more broadly defined open Internet.1
Specifically, we submit that beyond redefining network
neutrality, we must connect issues usually dealt with separatelyissues that are actually a subset of one overarching concern: the struggle for a democratic Internet. We
conclude with an exploration of the social and political
impacts of this broader conception of network neutrality.
These recommendations, we argue, provide a proactive
foundation for creating a more open and participatory
Internet. This project builds upon an earlier formulation
that we referred to as the New Network Neutrality.
W H Y N E T N E U T R A L I T Y M AT T E R S
Network neutrality helps ensure that telecommunication infrastructures remain dumb, delivering content and
services equally in a best-effort manner that treats data/
content delivery equitably. This best effort entails packets
being delivered in a first-in first-out method at the maximum speed possible given network constraints. Under a
framework of network neutrality, network operators do not
decide what content users can access and cannot impede
the flow of or give preferential treatment to particular
kinds of content. The loss of network neutrality provisions,
in effect, removes a crucial safeguard and increases the likelihood of a discriminatory telecommunications system.
A largely straightforward question of how the network will be operated has been rendered unnecessarily
opaque by some of the actors in these debates. In particular, public relations representatives of the phone and cable
companies who stand to gain the most from an Internet
stripped of net neutrality protections devoted considerable
resources toward averting consensus while maintaining a
faade of a debate over democratic requirements for an
open Internet. Some industry-funded astro turf groups
have recast the debate as one over government regulation.
Christopher Wolf, Co-Chair of HandsOff.org, claimed
that There is no established definition for the concept
of net neutrality and that Such government control
over the evolution of the Internet is unprecedented.2
Similarly, former Congressman Dick Armeys organization
FreedomWorks advocates for such a laissez faire regulatory
approach.3 Many of these claims against regulation ignore
12

December

2008

the fact that the federal government developed, funded,


and directly managed the Internet for more than a quarter
of a century (until US officials handed over network control to the private sector in the mid-1990s), and continues
to actively regulate and subsidize the Internet. Moreover,
many of these self-defined anti-regulation organizations
appear oddly complacent toward heavy-handed private
control over the Internet.
While groups like HandsOff seem driven by a categorical opposition against all governmental interventions,
public interest advocates, on the other hand, tend to
stress openness as their main concerns. Many groups and
individuals within the Internet freedom coalition have
sought net neutrality as an end in itself, while others have
cautioned that net neutrality does not fully restore common carriage and should not be seen as a silver bullet. Josh
Breitbart makes the important point that net neutrality is
actually a retreat from earlier ideas such as open access and
common carriage, which were US law prior to Brand X:
Net neutrality is actually a retreat from open
access, which is what we had before the Supreme
Courts Brand X decision. Open access applied to
the Internet when we were using dial-up and it was
classified as a telecommunications service like the
telephone. With the telephone, that means owners
of the lines cant prioritize their customers calls over
those of their competitors (net neutrality or, as its
known in the phone world, common carrier), but
it also means they have to lease their lines to other
phone service providers (open access). Open access
is how you can have real competition without having a dozen different wires running under your street,
through your backyard, and into your house.4
As Breitbart suggests, network neutrality is inextricably
linked to principles of common carriage, a bedrock principle of telecommunications policy for nearly 100 years
that mandated non-discriminatory service.5
Making sense of what is at stake in current net neutrality debates requires examining key antecedents in
telecommunications history. In the following section, we
examine the historical context and the current parameters of the debate. This historical context, we argue, is
necessary to begin imagining alternative trajectories for
internet policy.
H I S TO R I C A L OV E RV I E W
Prior to common carriage laws, telecommunication
operators were able to abuse their market position. During
the Civil War, Western Union controlled telegraph trunk

December

2008

lines across the country and gradually achieved near


monopolistic dominance by buying up competing companies and actively undercutting congressional and popular
support for constructing a rival postal telegraph system. As
its network expanded in the 20th century, Western Union
focused on serving business clients while pricing potential
competitors out of specific geographic markets and ignoring social obligations, such as universal service. Based on
Western Unions business model and the larger political
and regulatory environment, there was little incentive to
create innovations that could have made access affordable
for average citizens. With the rise of telephony and early
build out of telephone networks during the early 20th
century, public service protections were introduced into
the telecommunications regulatory structure, including
common carriage.
For decades, telephone network operators were considered natural monopolies. The biggest monopoly
(and largest corporation in the world) was the American
Telegraph and Telephone Company, which, until its
forced 1984 breakup into the baby bells, dominated the
telecommunications industry. Given its monopoly privileges, it was mandated as a basic public service that AT&T
could not discriminate against other carriers using its lines.
Leading up to the Modified Final Judgment that broke up
the Bell system, Judge Harold Green in 1982 maintained
that phone companies should not sell information. At the
time, former FCC commissioner Nicholas Johnson argued
that allowing phone companies to provide both conduit
and content would hurt both businesses and consumers
instead of providing the channels of communication for a
democratic society. Johnson argued that the phone companies drive to get into the information selling business
was the No. 1 public policy issue confronting our nation.
Arguing that they already profit from both ends of the
process, he worried that telecom providers charge us for
getting information out of the system and they charge the
supplier for putting it in.6 Commissioner Johnsons fears
became even more warranted after the 1984 Cable Act
and the 1996 Telecommunications Act, which, respectively, allowed cable to remain a closed system and to
become an Internet service provider.
The Telecommunications Act of 1996 was the first
major overhaul of the landmark 1934 Communications
Act and the first comprehensive attempt to reform US
media policy for the digital era. This complex and farreaching legislation replaced structural regulation with
market incentives for telephony, radio, broadcast television, cable television, and satellite communications.
Hailed as an effort to unshackle market forces and sold
with the promise that deregulation leads to enhanced
competition, the bill has instead led to unprecedented

J O U R N A L O F I N T E R N E T L AW

telecommunications conglomeration, lessened consumer


protections, and decreased ownership diversity. However,
taking for granted the historical importance of common
carriage in curbing market excesses, even the deregulatory
thrust of the 1996 Telecom Act left the principles of nondiscrimination intact.
This changed with the June 27, 2005, Supreme Court
Brand X decision and subsequent August 5, 2005, FCC
decision to deregulate carriage. Culminating after a long
legal fight between cable companies (like Comcast and
Time Warner) and independent ISPs (like Earthlink and
Brand X) over whether cable operators should be required
to sell access to their networks to potential broadband
service provision competitors, the Brand X decision was
significant for essentially deregulating broadband. The
Supreme Courts 6-3 decision favored the National Cable
and Telecommunications Association (NCTA), the principal trade association of the cable television industry, by
overturning an earlier appellate court decision and affirming the FCC classification that cable broadband was an
information service instead of a telecommunications
service, thus exempting cable companies from common
carriage laws.
This seemingly minor turn of phrase meant that cable
providers did not have to share their infrastructure with
competitors. Together with the subsequent FCC decision
to extend this exemption to phone companies (ostensibly
to provide a level playing field among market players), this
court decision removed safeguards and created the potential for access restrictions to non-preferred content. Many
public interest advocates pointed out how this decision
countered 100 years of telecom policy and risked changing
the open and non-discriminatory nature of the Internet
while creating a new class of potential gatekeepers.
MONOPOLY POWER AND
CONTENT DISCRIMINATION
The history of content control goes back centuries
and bears mentioning given parallels to the outcomes that
network neutrality advocates fear today, particularly when
looking at the history of the US postal system. For example,
analogies can be seen with the abortive attempt in the
early 1790s to admit only certain newspapers into the mail.
Congress rejected this policy when it enacted the Post
Office Act of 1792, which put into law principles of nondiscrimination. However, content restrictions persisted,
such as postal administrators blocking the dissemination
of books via mail, which they argued were too bulky, and
southerners after 1835 blocking the circulation of information on slavery.7 Although varying degrees of content
discrimination have persisted, in general the opportunity
13

J O U R N A L O F I N T E R N E T L AW

for anyone to send anything anywhere without constraint


or discrimination was a fundamental assumption of this
early US communications system. Alexis de Tocqueville,
who credited newspapers and other information delivered
via the post as greatly responsible for Americas thriving
democratic culture, praised the US system.8
This openness was periodically challenged, particularly by the monopolistic telegraph industry, which abused
its market power. Paul Starr notes historical parallels with
contemporary telecommunications marketplaces in which
incumbents dominate networks to exploit their existing
position rather than innovate and spend little money
on research and development, often investing more in
politics than in technology. Similar market conditions
exist today. Once again, first-mile telecommunications
are heading toward near-monopoly status. This time,
however, a crucial safeguard is missing; the Internet is no
longer classified as a telecom service and is at risk in being
transformed into a cable television business model.9
Despite network operators assurances to the contrary,
over the past several years there have been ominous
glimpses of what a non-neutral network might look like.
In 2004, North Carolina ISP Madison River blocked DSL
customers from using its rivals (Vonage) VOIP telephony
services. In 2005, the Canadian telecom corporation,
Telus, blocked users from accessing a pro-union Web
site during a labor dispute. In 2006, AOL Time Warner
blocked a mass email campaign from its customers that
opposed AOLs proposed tiered email system. In 2007,
AT&T apparently censored a Webcast of the rock band
Pearl Jams anti-Bush political commentary. Also in 2007,
Verizon was found blocking the pro-choice organization
NARALs text messages. In 2008, Bell Canada was caught
throttling third-party DSL providers P2P traffic. Perhaps
best exemplifying the potential for abusing net neutrality has been Comcast, whose practices of blocking traffic
associated with Bit Torrent, a peer-to-peer file-sharing
system, were exposed in 2007. These are just a few of the
more egregious infractions against net neutrality.
P R E V I O U S L I T E R AT U R E
Three waves of scholarship addressing the net neutrality debate can be discerned thus far. Although anticipated by earlier debates, the first phase was marked by
Timothy Wus initial formulation of network neutrality
in his seminal 2003 work, Network Neutrality, Broadband
Discrimination, where he forwarded the idea that network
architectures should be neutral purveyors of data.10 The
debate simmered among a relatively small group of commentators until the Supreme Courts pivotal Brand X
decision, which catapulted net neutrality to a new level of
14

December

2008

urgency as the prospects of tiered Internet services paralleling a cable television business model became a distinct
reality.11
The Brand X decision ushered in a second wave of
scholarship that was remarkably cautious given the stakes
involved. For example, Eli Noam has suggested a Third
Way for net neutrality limited to Last Mile concerns.12
Christian Sandvig was quick to discount some arguments
posed by network neutrality advocates, suggesting that
many aspects of net neutrality smacks of an old debate,
evidenced by principles laid out by Ithiel de Sola Pool
decades earlier. Noting that network neutrality has never
been the norm given that all Internet providers have
discriminated against certain types of content to some
extent, Sandvig called for establishing a set of normative guidelines to distinguish acceptable types of traffic
shaping.13 Less common in this second wave were articles
that staked out a position boldly calling for mandated net
neutrality.14
Currently, we have reached a new phase of the
debate, one that places net neutrality provisions in a
state of uncertainty. Although prospects seem less dire
than when we began working on this issue in 2005-2006,
net neutrality protections are still not codified into law.
However, even as scholarship has become less complacent
toward the loss of net neutrality, we submit that now is
precisely the moment that we should be aiming beyond
mandated net neutrality for more encompassing safeguards
to ensure an open Internet. Much of the existing scholarship and commentary fails to sufficiently emphasize the
import of normative principlesprinciples regarding the
role of the Internet in a democratic society and the debt
that Internet providers owe to the public. When considering the fact that the four Bell companies earn roughly
$14 billion every year from access to Internet content and
applications in addition to $20 billion a year in direct
access fees from broadband Internet subscribers and when
taken in the context of the enormous tax subsidies and
other benefits that telecom corporations receive from
public entities, this debate should focus more on the social
contract between telecom network operators and the public. These kinds of social contract debates often present
themselves during critical junctures and periods of media
crisis.15 The fact that network neutrality is a normative
principle is far too often overlooked. Industry attempts
to reframe the debate, growing technological complexity,
and shifting allegiances among competing actors artificially sunder democratic Internet principles that should
be considered together.16
Contrary to these general trends lies a neglected tradition located in scholarship that addresses more normative
concerns like open architecture, open access, and online

December

2008

ethics.17 Wu offers a short list of network neutrality rules


that would prohibit carriers from discriminating content.
Similarly, Benklers Wealth of Networks advocates for a
commons-based policy orientation. Along with Lessig
and others, this approach is aligned with the notion of
Coopers open architecture.18 Drawing from the research
of Yochai Benkler, Mark Cooper, Lawrence Lessig, Tim
Wu, and others, we envision a more open and participatory Internet. Frequently referred to as a commons-based
approach to the management of communications systems,
this model emphasizes cooperation and innovation as
opposed to privatization and enclosure. Given that all
technology is inscribed with social values that foreclose
certain possibilities while encouraging others, emphasizing these linkages illuminates what is at stake with network neutrality and situates this debate within a larger
vision of Internet openness. We sit at a critical juncture
for Internet policy; opportunities now abound that soon
will disappear.
C U R R E N T S TAT E O F A F FA I R S
While net neutrality helps prevent many of the
worst market excesses, it does little to ameliorate some
of the systemic problems that necessitate it. Media
conglomeration and the attendant lack of diversity of
ownership and perspectives provide one focal point
for discussing network neutrality.19 From the reemergence of telecommunications giant AT&T to current
efforts by FCC Chairman Kevin Martin to relax media
ownership restrictions, fewer players are gaining massive market share, creating increasingly vertically and
horizontally integrated corporations with the potential
to dominate entire market sectors.20 By many measures,
the current FCC regulatory environment fails to spur
technological innovation and has retarded expansion of
digital inclusion efforts.21 Instead, the FCC has fostered
a decades-long market environment fraught with pricing
and geographical discrimination as well as overpriced,
substandard telecommunications services.22
Exacerbating difficulties in these crucial media policy
areas are state and national telecommunications laws that
slow innovation and competition in broadband services,
thus creating an environment of digital exclusion. To
date, more than a dozen states have passed laws that in
some way limit competition and prevent innovation in
business models, public investment, and public-private
partnerships.23 At the national level, everything from
local control over local rights-of-way to consumer protections would be undermined by pending legislation.
While yesteryears newspapers and todays Internet
are quite different media, their social functionality

J O U R N A L O F I N T E R N E T L AW

within civil society is remarkably similar. Whereas the


unrestricted transport of newspapers via the postal service has long been protected and subsidized, today ISPs
are proposing to have discriminatory power over social
networking applications that use their networks. Using
the postal service, anyone can send packets first-class,
second-class, third-class, parcel post, overnight, etc.
However, when one sends a packet, it will be handled
in a first-in-first-out manner within the chosen service
without regard to the type of packet being sent. Likewise,
network neutrality incorporates strong civil rights protections simply by mandating a neutral and non-reactive
transport medium.
A related issue underlies concerns over surveillance.
Recent endeavors to surveil network traffic encroach
upon users rights to privacy, creating a panoptic environment that undermines civil society, creativity, and public
dialogue. Current law enforcement efforts should err on
the side of maintaining network neutrality, yet often
mandate data collection of user information that demonstrates both a lack of understanding of the current state of
technology and, in actuality, undermines long-term lawenforcement goals (as discussed later in this article). The
Communications Aid to Law Enforcement Act (CALEA)
is just one example of significant risks posed to Internet
freedom and, ironically, long-term law enforcement.
ABUSE OF MONOPOLY MARKET POWER
The rise of telephony over the past century suggests
that our current path has been tried before. As Paul Starr
writes in The Creation of the Media, From 1894 until
1907. . . the market broke open with a surge of independent commercial and nonprofit cooperative telephone
enterprises.24 AT&T and the Bell system, however, as
the primary owner of telephone long-distance service,
often refused to interconnect these independent commercial and nonprofit cooperative enterprises wherever
they were in competition for local phone customers.
Instead, AT&T used its long-distance monopoly to open
3,500 new exchanges in smaller communities of less
than 10,000 people between 1894 and 1907. As Starr
sums up:
The Bell-independent rivalry at the turn of the
century led to the same breakneck extension of
networks that had characterized the early telegraph
industry around 1850. . . prices for telephone service
fell sharply. Independent phone companies generally offered lower rates than Bell, and though Bell
cut its rates everywhere, they were lower where it
faced a rival.25
15

J O U R N A L O F I N T E R N E T L AW

At least until the 1913 Kingsbury Agreement, AT&T


interconnected with Independents when it suited its
needs; however, it preferred to buy out or quash these
competitors. Today, in the wake of Brand X, a market and
regulatory environment has been recreated that eliminates
independent companies and allows backhaul owners to
engage in similar anti-competitive practices.
In 1907, AT&Ts new president, Theodore Vail,
publicly declared that telephone service should be, in
essence, a unified, interoperable, neutral network. AT&T,
through the judicious use of governmental regulationfor
which AT&T often directly lobbiedwas able to create a
national interconnected telephone network and grow its
market share dramatically during the first three decades of
the 20th century (to 66 percent in 1920 and 81 percent
in 1932), crushing the home rule telephone movement,
thus ensuring decades of market dominance until the 1984
divestiture. The public statements of todays telecommunications leaders are explicitly interested in devising
ways to close off their networks, maximize their billable
minutes, and create new avenues for extracting additional
fees for service quality, non-interference, and non-discrimination. This sensibility is best exemplified by AT&T
and statements made by its CEO, Ed Whitacre:
I think the content providers should be paying for
the use of the networkobviously not the piece
from the customer to the network, which has
already been paid for by the customer in Internet
access feesbut for accessing the so-called Internet
cloud . . . If someone wants to transmit a high quality service with no interruptions and guaranteed
this, guaranteed that, they should be willing to
pay for that . . . They shouldnt get on and expect
a free ride.26
An important lesson is that AT&T gained its prominence not by any superior business model alone, but
through governmental regulation, predatory pricing,
buying up competition, centralizing network control, and
a dedication to creating and controlling a nationally interconnected network. Today, in much the same way it undermined the home rule telephone movement 100 years
ago, AT&T is again attempting to leverage its network
ownershipthis time targeting not only telephone services
but also all Internet-mediated communications, including
data services, streaming audio and video, and television.
FCC REGULATORY ENVIRONMENT
Network neutrality protections treat the question
of access as a critical element in determining whether a
16

December

2008

network is being operated in an open manner. Bottlenecks


to network access undermine the types of services offered,
create artificial scarcity, and lead to increased pricing and
lowered quality of service. This is exemplified by current
national policy surrounding the licensure of the public
airwaves.27 Well more than 99 percent of the public airwaves are either reserved for governmental use or licensed
to private companies.28 Even though the tiny sliver of
so-called unlicensed frequencies has generated enormous
economic activity and innovation, everything from WiFi
devices to baby monitors, radio phones, garage-door
openers, and microwave ovens coexist within these rare
frequencies.29
The FCC has continued to privilege a model for
licensure that allows only a single entity to broadcast
on a given swath of spectrum, often at a specific power
level and geographic location. While digital technologies
have radically transformed almost every aspect of current
society, our licensure regime is predicated on use of the
public airwaves as if we were still using 1920s and 30s
technologies. Whether one looks at the debate over lowpower FM radio licensure, interference temperature, or
unlicensed devices in unused television broadcast bands,
the story is invariably the same: Incumbent interests
already invested in licensed frequencies seek to prevent
competition by maintaining the licensure status quo and
thereby dramatically slowing down change or stopping it
altogether.
Given the accumulating evidence for broadband
connectivitys importance for economic development,
purposefully limiting access to the necessary tools to
build data communications networks is a disservice to
the general populace.30 Today, most wireless broadband
providers are forced to use only a handful of unlicensed
frequencies, creating a scarcity of capacity in dense
urban areas. Meanwhile, rural areas are often completely
neglected by broadband providers. Opening up large
swaths of unlicensed frequencies would not only help
meet current demand but also provide ample spectrum for
future technologies such as cognitive and software defined
radios. Yet proceedings to open up additional bands such
as 3650-3700MHz or to open up bands to more users and
unlicensed devices in unused 700MHz television broadcast frequencies continue to stagnate.
These same problems exist with other telecommunications media. Classical economics dictate that a glut
of supply should lower pricing. However, US broadband
pricing remains exceptionally high despite the open secret
that a majority of fiber infrastructure in the United States
is dark and remains underused. Information on where
this dark fiber exists and how much is available is considered a trade secret. In his book Broadbandits: Inside

December

2008

the $750 Billion Telecom Heist, Om Malik discusses the


enormous infrastructure overbuild of the late 1990s.31 The
reverberating effects from this $750 billion market failure
are still hindering US broadband development today.
Meanwhile, the FCC has in many cases systematically
removed the few remaining checks and balances protecting US residents from corporate malfeasance and market
excesses.
US BROADBAND PENETRATION RATES
Numerous states have passed laws restricting municipal entry into broadband service provision. Prior to 2005,
14 states created barriers to municipal broadband service
provision, which ranged from outright bans on public
utility districts providing retail telecommunications services to taxes on telecommunications services provided
by public entities (but not private providers) to increase
their prices.32 In response to direct lobbying by telecommunications incumbents, more than a dozen states have
passed regulations restricting competition in this market
sector.33 As the city of New Orleans discovered during
Hurricane Katrina, these state laws often force municipal
entities to spend crucial resources on making their networks worse.
Stagnation of US broadband penetration rates relative to a growing number of industrialized nations is due to
the combination of the aforementioned factors. While the
nation continues to lag further behind, this suboptimal
state of affairs is continually worsened by official comments, reports, and protocols that purposefully confuse the
issue and hide the extent of the problem.34 For example,
on July 26, 2006, the FCC released its most recent figures
on high-speed services for Internet access. In previous
years, the FCC had been lambasted for stating that 99 percent of the population had access to broadband services.
Numerous experts provided feedback on how the data
collected by the FCC could be improved so that its report
would provide more useful information, such as collect
information based on census track, disaggregate satellite
and other services, and make explicit the speeds of the
services provided.35 Yet, the 2006 report does almost nothing to address the fundamental concerns raised. Instead,
the FCC chose to wordsmith a solution that ignored the
requested feedback, leaving many issues without redress:
the inadequacy of the official definition for broadband
as 200kbps in a single direction; the severe limitations
of satellite as a medium for broadband service provision (in particular, speed limitations and latency, which
severely limit its utility for streaming, VoIP, and other
live services); and the lack of usefully disaggregated data.
By systematically suppressing competition and erecting

J O U R N A L O F I N T E R N E T L AW

numerous barriers to entry, the FCC and telecom incumbents have created an environment whereby substandard
and exorbitantly priced broadband service provision has
become the norm.
TOWA R D A N O P E N I N T E R N E T
We synthesize existing commons-based models to
create a more expansive standard of network neutrality
conducive to Internet opennessa model that runs counter to US phone and cable companies plans and challenges
the overly narrow parameters of current public interest
arguments. Discussion among pro-and anti-network neutrality camps often centers on the debate over quality
of services, bundling of services, and interconnection of
networks. At its core, the question is whether the Internet
should use an end-to-end infrastructure consisting of a
dumb network or whether a centralized infrastructure
should be used to inspect and shape network traffic based
upon its content, origin, and/or destination; thus, supporters of smart networks are often aligned with the antinetwork neutrality camp.36
Fundamental to smart networks is the idea that higher latency is not conducive to some services and applications. Thus, for example, VoIP (Voice over Internet
Protocol is packet-based telephony, a replacement phone
service) or streaming HDTV (High Definition TV) both
require low-latency and low-jitter throughput to be useful, whereas file transfers like Web surfing and email tend
to be relatively latency-agnostic and jitter-insensitive.
An ideal smart network would be able to distinguish
services and applications requiring low-latency and prioritize these network uses. The flip side is that low-priority network uses would find their latency increasing once
low-latency prioritization took place. A corollary of this
phenomenon is that latency is mainly an issue of network
capacity; with adequate capacity, packet prioritization
becomes a moot point. Thus, smart networks have the
potential to create a disincentive for system-wide capacity
upgrades.
For example, within Ethernet systems, network
neutrality might be circumvented through the use of
the 3-bit-wide Precedence section of the 8-bit Type
of Service field along with the existing 3-bit delay,
throughput, and reliability quality of service parameters.
Precedence is, for the most part, rarely used across most
public network infrastructure (though it is more prevalent
within private networks). While originally conceptualized
as a mechanism for determining the prioritization of traffic based on its import to network control (e.g., routine,
priority, immediate, flash, flash override, CRITIC/ECP,
Internetwork control, network control), it could also be
17

J O U R N A L O F I N T E R N E T L AW

used to discriminate content based upon purchase of premium, high-speed, or tiered services.
Additional complexity arises in the administration of
a packet-prioritizing network since this would necessarily involve some form of packet inspection (i.e., to identify what type of packet is being sent and its prioritization
level). Once system-wide prioritization levels are in place,
an incentive exists to create software to disguise data as
a higher priority form in order to speed its delivery. Thus, a
user might hide instant messaging data by using a program
that makes it appear to the network routers that these data
are VoIP packets; someone downloading MP3 files might
use an application that makes these data appear to be a
streaming audio file. Network providers, knowing that this
outcome is inevitable, would, in turn, need to do a deeper
packet inspection, further slowing network capacity as router CPU time is used to ensure that each packet is correctly
identified. A non-neutral network would create incentives
for non-high-speed content providers to use high-speed
content provision proxies to deliver content, creating
an entire market dedicated to concealment of data-location and counter-measures to prevent these initiatives by
network owners and those paying premium rates to avoid
content discrimination. Thus, without network neutrality,
a data-obfuscation arms race would certainly develop spanning all aspects of the networks infrastructure.
E X PA N D I N G T H E D E B AT E
In our view, the ways in which network neutrality has
been defined, with an emphasis on non-discriminating
wires and common carriage, are too limited in their scope.
Network neutrality advocates have been reacting to the
actions of incumbents and their lobbyists instead of formulating more proactive next steps. Using the current national
conversation as a springboard, we propose a far more encompassing perspective to help ensure network neutrality, one
that we believe will better enable the Internet to reach its
democratic and participatory potentials. Our recommendations go beyond questions of open access to consider the
broader contours of Internet architecture, including software, hardware, wireless/broadband infrastructure, ownership, economics, and open protocols and standards.
Our model for an open Internet contains 10 facets
that are necessary to ensure an interoperable, interconnected, non-discriminatory, global Internet. We assume
that competition is vital at all layers of Internet operations.
Without this competition, market capture through path
dependencya situation inherently detrimental to innovation and the best interests of network participantstends
to arise. While aspects of this analysis map onto the Open
Systems Interconnection (OSI) Reference Model, we also
18

December

2008

incorporate factors that help ensure a politically neutral


transport medium as well. In other words, neutrality is
not just a technical specification; it also facilitates a social
contract that supports equity and justice through data
communications. Given the shortcomings of traditional
neutral networking conceptualizations, this approach envisions a more democratic network infrastructure that:
1.
2.

Requires common carriage


Supports open architecture and open source driver
development
3. Maintains open protocols and open standards
4. Facilitates an end-to-end architecture (i.e., is based
upon a dumb network)
5. Safeguards privacy (e.g., no back doors, deep packet
inspection, etc.)
6. Fosters application-neutrality
7. Mandates low-latency and first-in/first-out (i.e., requires
adequate capacity)
8. Ensures interoperability
9. Remains business-model neutral.
10. Is governed by its users (i.e., is internationally representative and non-Amerocentric)
The following provides an initial skeleton for what
these 10 facets would entail; however, this is only a first
step toward achieving full implementation. Substantial
work is still required to flesh out these ideas.
RECOMMENDATION 1: REQUIRES
COMMON CARRIAGE
Common carriage ensures that network operators
lease their lines to all potential market players, including
municipalities, at market (wholesale) rates. Ideally, this
would include universal service provisions and service
level agreements. As has been seen repeatedly throughout
the history of transportation and telecommunications,
common carriage protects the general public against price
and geographic discrimination and other anti-competitive
business practices. Since 2000, the number of Internet
service providers has nearly halved (from 8,450 in 2001
to 4,417 in 2005). With the demise of common carriage
provisions resulting from the Brand X Supreme Court
decision, this number will continue to decrease.
RECOMMENDATION 2: SUPPORTS OPEN
ARCHITECTURE AND OPEN SOURCE
DRIVER DEVELOPMENT
Open architecture and open source driver development encourage a digital commons by keeping both the

December

2008

hardware itself and any hardware access layer(s) open. As


the open source movement gains ground (especially internationally) and hardware prices have plummeted, new
business models have arisen to promulgate market capture
and path dependence, creating potentials for secondary
network closure.37 Open architectures and access layers
help promote competition by creating opportunities for
new market entrants and rapid innovation of features and
functionality.
RECOMMENDATION 3: MAINTAINS OPEN
PROTOCOLS AND OPEN STANDARDS
Maintaining open protocols and standards helps
ensure free-flowing, non-enclosed Internet services. This,
in turn, facilitates innovation and widespread adoption of
technologies. With the growing pull toward proprietary
networking (especially within the wireless medium), it is
vitally important to prevent the so-called Balkanization
of the Internet. Protocols and standards are the building
blocks for everything from interoperability to end-to-end
connectivity.
RECOMMENDATION 4: SUPPORTS
AN END-TO-END ARCHITECTURE
End-to-end architectures (E2E) help remove
vulnerabilities to bottlenecks, gate-keeping, illegal
surveillance by telcos, etc. E2E helps speed network
throughput and increases network capacity while
lowering network equipment costs and supporting
peer-to-peer communications. An end-to-end architecture helps prevent both governmental and corporate
interference in network traffic, an outcome that is
especially important at a time when surveillance and
digital rights management concerns are increasingly
prevalent.
RECOMMENDATION 5:
SAFEGUARDS PRIVACY
Private networks do not privilege state security
imperatives that compromise individual privacy rights
and help ensure a non-discriminatory environment for
content access and information dissemination. Private
networking is essential since back doors and other
devices introduce both enormous security holes as well
as increased impetus for development and widespread
adoption of privacy software that hampers, over the longterm, legitimate law enforcement efforts. Privacy is also
essential for ensuring the continued expansion of online
business.38

J O U R N A L O F I N T E R N E T L AW

RECOMMENDATION 6: FOSTERS
APPLICATION-NEUTRALITY
With application neutrality, Internet television,
VoIP, and diverse operating systems and services run
unimpeded. Expected convergences in digital communications make this principle increasingly crucial to the
long-term growth and health of the Internet. Digital
Rights Management (DRM) considerations such as copyright also make this a critical facet for a more open
Internet. In much the same way that telephone systems
are neutral transport mediums for voice communications,
the Internet must remain free from discriminatory practices that privilege some applications, services, or features
over others.
RECOMMENDATION 7: MANDATES
LOW-LATENCY AND FIRST-IN/FIRST-OUT
Low-latency and first-in/first-out helps remove the
impetus for data packet and application discrimination by
requiring that a service providers profit margins adhere
to the fundamental basic corporate responsibility to provide adequate services to its customers. These mandates
help lower over-subscription rates, artificial scarcity, and
the hoarding of dark fiber assets by mandating adequate
capacity and providing incentive for network and capacity upgrades.
RECOMMENDATION 8:
ENSURES INTEROPERABILITY
Interoperability harmonizes different systems and
integrates foreign attachments. This is especially important to the continued global expansion of broadband
service provision. As Cooper points out, interoperability
lowers costs while increasing the collaborative potential
of the Internet. Interoperability is critical to ensuring that
the 80 percent of humanity who are not currently online
will be able to interconnect with next generation telecommunications infrastructures.
RECOMMENDATION 9: REMAINS
BUSINESS-MODEL NEUTRAL
A business-model-neutral infrastructure allows
for public players such as municipalities and nonprofits, as well as public-private partnerships and private
corporations, to provide Internet services. Too often,
competition is lessened, and the options for consumers to receive broadband services artificially limited,
by shortsighted rules, regulations, and laws. A neutral
19

J O U R N A L O F I N T E R N E T L AW

2008

network cannot exist when limited to specific business


models.

4. Josh Breitbart, Wheres My Open Access, Civil Defense blog (Aug. 7,


2006), http://www.freepress.net/news/16943.

RECOMMENDATION 10: IS GOVERNED


BY ITS USERS

6. Quoted in Denise Caruso, San Francisco Examiner (May 19, 1991),


retrieved from http://www.publicknowledge.org/blog/1556.

We recommend replacing and/or dramatically expanding control over important governance institutions like
ICANN in a way that internationalizes control over such a
vital global resource. The current US-controlled ICANN
model is unsustainable over the long term.39 Expanding
governance would also help remove artificial scarcity and
hoarding of IPv4 addresses. As Milton Mueller and others
have documented, control over global communications
networks and the Internet, in particular, has remained
Amerocentric.40 Moreover, purportedly representative
bodies like ICANN and the Regional Internet Registries
(RIRs) often appear to privilege industry interests.
CONCLUSION
We submit that the implementation of these 10 principles will create a more participatory Internet. On a fundamental level, an open system is key to network growth
and innovation. We acknowledge that our model does not
address all material inequities, such as digital divide and
lack of universal service issues, which, to be sufficiently
remedied, require a redistribution of critical resources.
Nor do we tackle some issues related to copyright, surveillance, and other contemporary political battles. However,
our recommendations, if enacted, could improve the
global deficit in Internet connectivity and help propel
the United States toward its goals of universal, affordable
broadband. These principles could help establish normative parameters to guide policy makers, both national and
global, in their quest to create a better Internet. Ideally,
these principles will be presented as a broadband democracy manifesto to be endorsed by members of Congress,
state legislatures, and political candidates. Although
piecemeal efforts are better than no movement at all,
only if approached in tandem will these steps constitute a
model for the Internet that is simultaneously open, democratic, and efficient.
N OT E S
1. This article builds on an earlier version written in fall 2008 published
as Sascha D. Meinrath and Victor W. Pickard, The New Network
Neutrality: Criteria for Internet Freedom, International Journal of
Communication Law and Policy, 12, 225-243 (2008).
2. See http://www.handsoff.org/blog/ Hands Off the Internet at one
point was spending roughly $400,000 a day on TV and newspaper
advertising to promote a position identical to corporate sponsors like
AT&T.
3. See http://freedomworks.org/.

20

December

5. Eli Noam, Beyond Liberalization II: The Impending Doom of Common


Carriage, Telecommunications Policy, Vol. 18, No. 6 pp.435-452 (1994).

7. Richard John, Spreading the News: The American Postal System from
Franklin to Morse (Cambridge: Harvard University Press 1998). We are
also indebted to Richard Johns personal communications regarding these
points.
8. Alexis de Tocqueville, Democracy in America (Anchor Books: Garden
City, NY, 1969).
9. Paul Starr, The Creation of the Media: Political Origins of Modern
Communications (Basic Books: New York, NY, 2004).
10. Timothy Wu, Network neutrality, Broadband Discrimination, Journal of
Telecommunications and High Technology Law (2), 141-179 (2003).
11. For a description of this case, see http://www.fcc.gov/ogc/documents/opinions/2005/04-277-062705.pdf.
12. Eli Noam, A Third Way for Net Neutrality, Financial Times (Aug. 29,
2006).
13. Christian Sandvig, Network Neutrality is the New Common Carriage,
Journal of Policy, Regulation, and Strategy for Telecommunications 9(2/3): 136147 (2007).
14. One notable exception is Bill Herman, Opening Bottlenecks: On Behalf
of Mandated Network neutrality. Federal Communications Law Journal,
59, 1, 107-159 (2007). Herman provided a comprehensive overview and
challenges Christopher Yoos argument that net neutrality provisions are
unnecessary. See Christopher Yoo, Beyond Network neutrality, Harvard
Journal of Law and Technology, Vol. 19 (2005).
15. For a similar argument, see Victor W. Pickard, Media Democracy
Deferred: The Postwar Settlement for U.S. Communications, 1945-1949
(PhD Dissertation, University of Illinois 2008).
16. Harold Feld, Thinking a Bit More Clearly on Grokster and Brand X,
WetMachine, Tales of the Sausage Factory, May 3, 2006, http://www.
wetmachine.com/totsf/item/500. See also Yochai Benkler, From Consumers
to Users: Shifting the Deeper Structures of Regulation Toward Sustainable
Commons and User Access, Federal Communications Law Journal, Vol. 52,
pp.561-579 (2000).
17. Mark Cooper, Open Communication Platforms: Cornerstone of
Innovation and Democratic Discourse in the Internet Age, Journal of
Telecommunications and High Technology Law, 1 (2003); E.W Comstock
& J.W. Butler, Access denied: The FCCs Failure to Implement Open
access as Required by the Communication Act, Journal of Communications
Law and Policy (Winter 2000); Cees Hamelink, The Ethics of Cyberspace
(London: Sage 2000).
18. Yochai Benklers (2006) The Wealth of Networks: How Social Production
Transforms Markets and Freedom (Yale University Press: New Haven, CT,
2006; Lawrence Lessig, The Future of Ideas: The Fate of the Commons in a
Connected World (Random House: New York, NY, 2001).
19. Ben Bagdikian, The Media Monopoly (Beacon Press: Boston, MA, 2000);
Robert McChesney, Rich Media, Poor Democracy: Communications
Politics in Dubious Times (The New Press: New York, NY, 1999).
20. Bruce Kushnick, The Unauthorized Bio of the Baby Bells & Info-Scandal
(New Network Institute: New York, NY, 1999).
21. Mark Cooper, Open Architecture as Communications Policy: Preserving
Internet Freedom in the Broadband Era (Stanford Law School Center for
Internet and Society: Stanford, CA, 2004).
22. H. R Slotten, Radio and Television Regulation: Broadcast Technology in
the United States, 1920-1960 (John Hopkins University Press: Baltimore,
MD, 2000).
23. See APPAs State Barriers to Community Broadband Services.
24. Paul Starr, supra n.9, at p.193.
25. Id. at 201-202.
26. Paul Taylor, AT&T chief warns on internet costs, http://www.ft.com/cms/
s/2/3ced445e-91c5-11da-bab9-0000779e2340.html.
27. Supra n.20.
28. Slotten, 2000.

December

2008

29. New America Foundation, The Citizens Guide to the Airwaves: A


Graphic Depiction of the Usesand Misusesof the Radio Frequency
Spectrum (Washington, DC, 2003).
30. W.H. Lehr, C.A. Osorio, S.E. Gillett, & M.A. Sirbu, Measuring
Broadbands Economic Impact, presented at the 33rd Research
Conference on Communication, Information, and Internet Policy
(TPRC). Arlington, VA, Sept. 23-25, 2005; S.E. Gillett, Municipal
Wireless Broadband: Hype or Harbinger?, presented at Symposium
Wireless Broadband: Is the U.S. Lagging?, Washington, DC, Oct. 26,
2005.
31. Om Malik, Broadbandits: Inside the $750 Billion Telecom Heist (John
Wiley & Sons, Inc: Hoboken, NJ).
32. James Baller, http://www.baller.com/pdfs/Baller_Proposed_State_Barriers.pdf
(accessed June 30, 2007).
33. Ironically, these are often the same companies, such as AT&T, which are
lobbying against similar regulations at the national level.
34. The OECD puts the US in 12th place among its 31 members as of
December 2005; the 2005 statistics from the ITU places the US in 16th
place. In 2008, the OECD still ranked the 15th for overall nationwide
broadband penetration. This report runs contrary to recent governmental assurances regarding the rosy outlook on the state of broadband in
the United States. See National Telecommunications and Information
Administration, Networked Nation: Broadband in America, Jan. 30,
2008. This report concludes that a reasonable assessment of the available data indicates that President Bushs 2004 objective of affordable

J O U R N A L O F I N T E R N E T L AW

access to broadband for all by 2007 has been realized to a very great
degree.
35. Derek Turner, Broadband Reality Check: The FCC Ignores Americas
Digital Divide, available at http://www.freepress.net/docs/broadband_
report.pdf.
36. J.H. Saltzer, D.P. Reed, & D.D.Clark (1984), Transactions on Computer
Systems 2, 4 (Nov. 1984), pp.277-288; David Isenberg, The Rise of the
Stupid Network, retrieved Dec. 1, 2006 from http://www.rageboy.com/
stupidnet.html.
37. See Christian Sandvig, DavidYoung, & Sascha D. Meinrath, Hidden
Interfaces to Ownerless Networks, presented to the 32nd Conference on
Communication, Information, and Internet Policy (TPRC), Washington,
DC, Sept. 2004.
38. Gerald Brock, The Second Information Revolution (Harvard University
Press: Cambridge 2003); Ursula Huws. The Making of a Cybertariat: Virtual
Work in a Real World (Monthly Review Press: New York, NY, 2003).
39. Victor W. Pickard, Neoliberal Visions and Revisions in Global
Communication Systems from NWICO to WSIS, Journal of
Communication Inquiry, 31 (2), 118-139 (2007).
40. Milton Mueller, Ruling the Root: Internet Governance and the Taming
of Cyberspace (MIT Press: Cambridge, MA, 2002). See also A. Mattelart,
Mapping World Communication: War, Progress, Culture (University of
Minnesota Press: Minneapolis, MN, 1994); Paschal Preston, Reshaping
Communications (Sage Publications: Thousand Oaks, CA, 2001).

21